PROJECT PROPOSAL 
Jacqueline Signorini & Patrick Greussay

 

Abstract

When the going get rough, and if we are lucky, here comes an ambulance with experienced paramedics: we get 15mg of morphine, the pain is alleviated but remains sufficiently high that transportation is still not possible. Paramedics are reluctant to add more morphine, something else is needed, but at which dosage? This is a systemic question involving molecular and physiological processes. What to do? Ask the Medical Emergency Embedded Active Knowledge Base (MEEAKB) via its interface. The answer will be of course Laroxyl with a dosage of no more than 50mg.
Our proposal is a first specification of this integrated tool along with the previous example handled at the molecular, cell or organ level. Plus a first step towards its question-answering interface.

I intend to propose a model for the building of an embedded active knowledge base as a tool for medical emergency diagnosis and prescriptions.
Let me explain what MEEAKB is about:

 
I illustrate what I mean with a simple scenario:

A gentleman over sixty suddenly gets into acute muscular pain. Even morphine cannot entirely overcome the pain, so that additional care must be given before the transportation to a hospital. Our gentleman has family antecedent in chronic orthostatic hypotension, is already diagnosed with massive decalcification and is known for negative response to anti-inflammatory medications.
                          Our knowledge base must work as follows:
More morphine is out of the question because of hypotension issues. Anti-inflammatories are very dangerous. To get over the pain to the point that the patient can be moved to the hospital, the doctor should prescribe a tricyclic antidepressant, particularly amitriptyline, known for its analgesic effect in neuropathic pain: Laroxyl (amitriptyline) with Tramadol. Which dosage? The answer must be quick: there is an obvious risk of seizure during coadministration of Tramadol with selective serotonin reuptake inhibitors and of any substance that can reduce the seizure threshold. These agents may have additive effects during coadministration. Besides, linked to blood pressure problems, complications with kidneys and liver are likely to occur. The knowledge base must therefore propose Laroxyl with 50mg for the next 20 minutes.

 

To be able to suggest the immediate treatment, the knowledge base must contain active knowledge of processes at molecular, organic, pharmacological levels along with their integration into a super-process and its interface to help the emergency medical person.

To exist, our base must include:

  1. Detailed specification, at industrial strength, of medical processes.
  2. Integrated simulators for the physiological processes into a constraint-based deduction system and suggestion interface.
  3. Robustness at every level of the medical knowledge involved along with enough openness to accept a parameter-based reconfigurable suggestion-system.

At this moment, I am able to give a robust and standardized set of specifications for processes such as: deficiency in coagulation (biochemical level), deficiency in functional cell cluster in the liver (sub-organ level), deficiency in basic filtering in the kidney (organ and systemic levels).
I propose:

Moreover, I am now consulting the VisualMED Solutions company, based in Montreal, which has developed a repository-based clinical toolset and integrated electronic patient record. This company is of particular relevance, due to its real world experience with medical databases, and the feedback it has received from hospitals and medical personnel.

References
1.  J. Signorini & P. Greussay, A Case Study of Object-Oriented Bio-Chemistry: A Unified Specification of the Coagulation Cascade, Genome Informatics, Vol.14:1, 2003
2.  J. Signorini & P. Greussay, Object-Oriented Specification of Complex Bio-Computing Processes: Blood Clotting as a Case Study, LNCS Vol. 3141, 2004
3. J. Signorini & P. Greussay, Object-Oriented Wound Healing in the Liver: a Class-Structured View of Fibrogenesis and a Glimpse of its Evolution, ACM SIGAPP Symposium on Applied Computing, Santa Fe (USA), 2005
4. J. Signorini & P. Greussay, Silent Killing: an Object-Oriented View of Hypertension and Kidney Failures, IEEE 7th Symposium on Bioinformatics, Harvard Medical School, 2007